The algorithm extracts shared variation from a collection of data sets using regression models.
cocoreg is an R-package for extracting shared variation in collections of datasets using regression models. The current stable release is available in CRAN:
The algorithm is described in the paper:
Using regression makes extraction of shared variation in multiple datasets easy: Jussi Korpela, Andreas Henelius, Lauri Ahonen, Arto Klami, Kai Puolamäki, Data Mining and Knowledge Discovery, 2016. URL: http://dx.doi.org/10.1007/s10618-016-0465-y
The authors' version is available in this repository as cocoreg_plain.pdf. The final publication will be available at link.springer.com.
A minimal usage example:
library(cocoreg)dc <- create_syn_data_toy()ccr <- cocoreg(dc$data)shared.by.all.df <- variation_shared_by(dc, 'all') #only on synthetic datasetsggplot_dclst(list(observed = dc$data, shared = shared.by.all.df, cocoreg = ccr$data))
library(reshape) #importing from namespace does not work as expectedggcompare_dclst(list(shared = shared.by.all.df, cocoreg = ccr$data))
The most important functions in cocoreg are:
cocoreg() which extracts shared variation from a collection of datasets
Functions to visualize output such as
ggcompare_dclst() for lists of data collections,
ggplot_dflst() for lists of data.frames (i.e. one data collection) and
ggplot_df() for a single data.frame (a dataset)
Install the release version from CRAN:
Or the development version from GitHub:
Fixed documentation (NAMESPACE exports, spell check) and added README.md figures as part of the package in "man/figures".
Initial CRAN release upon article publication.